Graduate Student Researcher

I'm a third year computer science M.S./Ph.D. candidate working with David Jensen in the Knowledge Discovery Laboratory at UMass Amherst. My research focus is on how to bridge the gap between mechanistic models and machine learning, enabling AI-assisted scientific discovery and explainable AI. This work synthesizes several subfields, including causal graphical models, probabilistic programming, deep learning, and reinforcement learning. When I'm not reading/writing/talking about data, you're likely to find me lost in the woods with my puppy Mira.

I'm very excited to be spending the Summer of 2019 as a visiting researcher at Invitae, where I'll be working on causal modeling for genetics applications.


Measuring and Characterizing Generalization in Deep Reinforcement Learning. Sam Witty, Jun Ki Lee, Emma Tosch, Akanksha Atrey, Michael Littman, David Jensen (2018). arXiv preprint arXiv:1812.02868 (Workshop version published at the NeurIPS Workshop on Critiquing and Correcting Trends in Machine Learning.)

Causal Graphs vs. Causal Programs: The Case of Conditional Branching. Sam Witty, David Jensen (2018). Proceedings of the First Conference on Probablistic Programming.

Belief-Space Planning for Automated Malware Defense. Justin Svegliato, Sam Witty, Amir Houmansadr, Shlomo Zilberstein (2018). IJCAI Workshop on AI for Internet of Things.


Teaching and Mentorship

  • I'm the teaching assistant for CS348, Umass' upper-level undergraduate course on data science. (Spring, 2019)

  • I gave a guest lecture on deep learning for CS589, Umass' Masters course on machine learning. (February 15, 2018)

  • I mentored Catherine Chen, a visiting undergraduate researcher sponsored by the NSF's REU program. (Summer, 2017)